ResCap-DBP: A Lightweight Residual-Capsule Network for Accurate DNA-Binding Protein Prediction Using Global ProteinBERT Embeddings

2025-07-27

AI Summary

To address the time-consuming and costly experimental identification of DNA-binding proteins (DBPs), this paper proposes ResCap-DBP, a lightweight residual capsule network that learns discriminative features directly from raw amino acid sequences. Methodologically, it innovatively integrates residual learning, dilated convolutions, and a one-dimensional capsule network: residual learning mitigates gradient vanishing, while the capsule architecture—via dynamic routing—captures high-order spatial relationships among sequence features. Additionally, ProteinBERT-derived global embeddings are incorporated to enhance generalization. Evaluated on benchmark datasets PDB14189 and PDB1075, ResCap-DBP achieves AUC scores of 98.0% and 89.5%, respectively; on independent test sets PDB2272 and PDB186, it attains AUCs exceeding 83.2%. It significantly outperforms state-of-the-art methods, demonstrating superior sensitivity and specificity. This work provides an efficient, robust deep learning solution for accurate DBP prediction under both small- and large-sample scenarios.

Technology Category

• Machine Learning: Deep Neural Architectures and Foundation Models
• Search and Optimization: Learning to Search
• Intelligent Robots: Learning & Optimization for ROB

Application Category

• Graph Algorithms and Modeling for the Web: Graph neural networks and deep learning approaches for Web-related graphs
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Web Mining and Content Analysis: Robustness and generalizability of Web mining methods

Abstract

DNA-binding proteins (DBPs) are integral to gene regulation and cellular processes, making their accurate identification essential for understanding biological functions and disease mechanisms. Experimental methods for DBP identification are time-consuming and costly, driving the need for efficient computational prediction techniques. In this study, we propose a novel deep learning framework, ResCap-DBP, that combines a residual learning-based encoder with a one-dimensional Capsule Network (1D-CapsNet) to predict DBPs directly from raw protein sequences. Our architecture incorporates dilated convolutions within residual blocks to mitigate vanishing gradient issues and extract rich sequence features, while capsule layers with dynamic routing capture hierarchical and spatial relationships within the learned feature space. We conducted comprehensive ablation studies comparing global and local embeddings from ProteinBERT and conventional one-hot encoding. Results show that ProteinBERT embeddings substantially outperform other representations on large datasets. Although one-hot encoding showed marginal advantages on smaller datasets, such as PDB186, it struggled to scale effectively. Extensive evaluations on four pairs of publicly available benchmark datasets demonstrate that our model consistently outperforms current state-of-the-art methods. It achieved AUC scores of 98.0% and 89.5% on PDB14189andPDB1075, respectively. On independent test sets PDB2272 and PDB186, the model attained top AUCs of 83.2% and 83.3%, while maintaining competitive performance on larger datasets such as PDB20000. Notably, the model maintains a well balanced sensitivity and specificity across datasets. These results demonstrate the efficacy and generalizability of integrating global protein representations with advanced deep learning architectures for reliable and scalable DBP prediction in diverse genomic contexts.

Problem

• Accurate prediction of DNA-binding proteins from sequences
• Overcoming limitations of costly experimental DBP identification
• Enhancing deep learning for scalable genomic protein analysis

Innovation

• Residual-Capsule Network for DBP prediction
• Uses ProteinBERT embeddings for global features
• Dilated convolutions prevent vanishing gradients